Project Summary/Abstract Cardiac arrest (CA) has an estimated annual incidence of 250-350,000 out-of-hospital and 250-750,000 in- hospital events in the U.S., with survival rates as low as 5% and 20% respectively. Anoxic brain injury remains a major health burden for those who survive to hospital discharge. These outcomes reflect a two-step injury process including a) whole body anoxia/isquemia and b) secondary reperfusion and inflammation injury following return of spontaneous circulation (ROSC). As secondary injuries are determined by the magnitude of ischemia during CA, the ability to ameliorate ischemia by improving resuscitation quality and increasing oxygen delivery in real-time is vital to reducing ischemic and subsequent secondary injuries. Current resuscitation methods only provide 25-30% of cardiac output and are limit by the inability of clinicians to deliver cardiopulmonary resuscitation (CPR) effectively up to 50% of the time. In response, the American Heart Association (AHA) has recommended two options of feedback to improve CPR quality: i) physiological or ii) non-physiological systems. Through a multi-site study, we have studied end tidal carbon dioxide (ETCO2) monitoring, and pioneered non-invasive brain monitoring of regional cerebral oxygenation (rSO2) using near infra-red spectroscopy as physiological feedback during CPR and demonstrated that they reflect two complementary aspects of CPR: i) circulation quality (ETCO2) and ii) brain/vital organ perfusion (rSO2). We have also shown that rSO2 exhibits a dose response relationship with survival and favorable neurological outcomes. In spite of these data and the AHA recommendations, the physiological target and the optimal feedback system during CPR to predict CA survival without neurological impairment remains unknown. In the present study, we propose to analyze data and blood samples collected from an ongoing 20 site observational study (AWAreness during Resuscitation [AWARE II]) of in-hospital cardiac arrests (IHCAs) to identify a brain resuscitation ?gold standard,? and then test the hypothesis that physiological feedback guided CPR using rSO2 or ETCO2 or a combination will perform better than non-physiological feedback CPR in predicting CA survival and neurological status through attenuation of brain injury and inflammation. The current application proposes three aims: For Aim 1, using prospective data from 500 IHCAs, we propose to identify the optimal physiological resuscitation target (rSO2 and/or ETCO2) in predicting CA survival and neurological status. For Aim 2, serum collected during CPR and in the post-CA period will be analyzed to measure biomarkers of brain injury and inflammation to determine the association between rSO2 and ETCO2 levels during CPR and markers of secondary injury. Having determined the optimal physiological target, Aim 3 will randomize 150 subjects to compare the impact of real-time physiological feedback CPR with non-physiological feedback CPR on ROSC, survival and neurological outcomes in a pilot randomized control trial.